JP6574014B2 - Video decoding method, video encoding method, and recording medium - Google Patents

Description

  The present invention relates to video coding, and more particularly, to intra prediction by applying an adaptive filter to a neighboring pixel value of a block to be predicted or a predicted pixel value of a current block. -Prediction) method and apparatus.

  In recent years, with the advent of digital TV and the like, technologies in the field of broadcast television and home entertainment have rapidly oscillated. Technologies in these fields are commercialized by standardization of video compression technology. For video compression, ITU-T (International Telecommunications Union-Telecommunication) H. The H.263 standard is widely used, and MPEG-4, the next standard of Motion Picture Experts Group (MPEG), is used for Internet-based video applications.

  H. After the completion of the H.263 standard, ITU-T VCEG (Video Coding Experts Group) We have raced a short-term goal to add additional features to the H.263 standard and a long-term goal to develop a new standard for low bitrate visual communication. In 2001, JVT (Joint Video Team) made by MPEG and VCEG experts was formed, and ITU-T H.264 is a new standard for coding video images by JVT. H.264 / MPEG-4 part 10 standardization work has been underway. H. The H.264 standard can also be called AVC (Advanced Video Coding). H. The technical goals of H.264 / AVC are significant improvements in coding efficiency, loss and error robust coding technology, network-friendly coding technology, low latency capability. , And exact match decoding.

  The peripheral pixels in the video have almost similar values. The same applies to 4 × 4 blocks or 16 × 16 blocks, which are the minimum block size of the H.264 / AVC standard. In this way, it is possible to perform prediction on a video using similarity such as a value between blocks and encode a difference from the original video. This is called intra-prediction, and the efficiency of video coding can be increased by intra-prediction.

  In addition, when executing intra prediction, a filter may be applied before executing intra prediction. Usually, H.M. When intra prediction is performed according to the H.264 / AVC standard, a filter is applied to a reference pixel value and the like, and then a value to which the filter is applied is used for intra prediction. However, in some cases, the performance of video coding can be higher when performing intra prediction without applying the filter than performing intra prediction after applying the filter.

  Accordingly, a method for determining whether or not to apply a filter can be proposed when performing intra prediction.

  A technical problem of the present invention is that intra-prediction is performed by applying an adaptive filter to neighboring pixel values of a block to be predicted by video encoding or a predicted pixel value of a current block. It is an object of the present invention to provide a method and apparatus. More specifically, after applying an adaptive filter (pre-filter) to the reference pixel value of the current block for performing intra prediction, the prediction is performed, and the predicted pixel value of the current block, etc. In addition, an adaptive filter (post-filter) is applied to calculate a residual signal.

  In one aspect, a method for performing intra-prediction is provided. The intra prediction execution method is determined to determine whether to apply a first filter to a reference pixel value based on information about a neighboring block of a current block, and to apply the first filter. A step of applying the first filter to the reference pixel value; a step of executing intra prediction on the current block based on the reference pixel value; and a step of executing the intra prediction based on information on the neighboring blocks. Determining whether to apply a second filter to a prediction value for each prediction mode of the current block predicted by the prediction execution; and, if it is determined to apply the second filter, The second value for the prediction value for each prediction mode of the current block. Applying a filter.

  The applicability of the first filter may be determined based on the prediction mode of the current block determined based on information on the neighboring blocks.

  The applicability of the first filter may be determined based on the size of the current block.

  The applicability of the first filter can be designated in advance based on the prediction mode of the current block and the size of the current block.

  The applicability of the first filter may be determined based on whether or not the peripheral block is intra-frame encoded or inter-frame encoded.

  The first filter may use at least one of a 3-tap filter and a 2-tap filter.

  The applicability of the second filter may be determined based on the prediction mode of the current block determined based on information on the neighboring blocks.

  The applicability of the second filter can be determined based on whether the peripheral block is intra-coded or inter-coded.

  The second filter may be applied to a predicted value of a pixel adjacent to the reference pixel value boundary.

  The second filter may use at least one of a 3-tap filter and a 2-tap filter.

  In another aspect, an encoder is provided. The encoder includes a processor and a memory connected to the processor and storing information for driving the processor. The processor refers to information on peripheral blocks of the current block as a basis. Whether to apply a first filter to a pixel value is determined, and if it is determined to apply the first filter, the first filter is applied to the reference pixel value, and the reference pixel value is Intra prediction for the current block is executed as a basis, and whether or not the second filter is applied to a prediction value for each prediction mode of the current block predicted by the intra prediction execution based on the information of the neighboring blocks is determined. , If it is decided to apply the second filter, each schedule of the current block is determined. Configured to apply the second filter to each mode of the predicted value.

  In yet another aspect, a decoder is provided. The decoder comprises a processor and a memory connected to the processor and storing information for driving the processor, the processor based on information on peripheral blocks of the current block as a basis for a first reference pixel value. If it is determined whether to apply a filter and it is determined to apply the first filter, the first filter is applied to the reference pixel value, and the current block is applied to the current block based on the reference pixel value. Intra prediction is performed, whether to apply a second filter to a prediction value for each prediction mode of the current block predicted by the intra prediction based on the information of the neighboring blocks is determined, and the second filter is If it is decided to apply, the second filter for the prediction value of each prediction mode of the current block Configured to apply.

  The encoding performance is improved by effectively predicting the luminance or color difference signal block to be encoded.

H. 1 is a block diagram of an H.264 / AVC (Advanced Video Coding) standard encoder. H. 2 is a block diagram of a H.264 / AVC standard decoder. FIG. It is a figure which shows the example which labeled the prediction sample (prediction sample) in 4 * 4 luma prediction mode. It is a figure which shows nine prediction modes in 4 * 4 luma prediction mode. It is a figure which shows an example of the method of applying a filter before intra prediction execution. It is a figure which shows one Embodiment of the intra prediction execution method using the proposed adaptive filter. It is a figure which shows the case where a filter is applied to a predicted value according to the intra prediction execution method using the proposed adaptive filter. It is a block diagram of an encoder and a decoder in which an embodiment of the present invention is implemented.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains can easily implement the embodiments. However, the present invention can be realized in various different forms and is not limited to the embodiments described herein. In the drawings, in order to clearly describe the present invention, portions not related to the description are omitted, and similar portions are denoted by similar reference numerals throughout the specification. In addition, descriptions of parts that can be easily understood by those skilled in the art even when detailed descriptions are omitted are omitted.

  Throughout the specification and claims, when a part “includes” a component, it does not exclude other components, unless stated to the contrary, unless otherwise stated. It can be further included.

  FIG. 1 is a block diagram of an H.264 / AVC (Advanced Video Coding) standard encoder.

  As shown in FIG. 1, the encoder includes two paths of data flow. One is a forward path, and the other is a reconstruction path.

First, the forward path will be described. An encoding process is performed on the input frame F _{n in} units of macro blocks. The macroblock has a size of 16 × 16 pixels in the original image. Intra-prediction or inter-prediction is performed on each input frame. Intra prediction performs prediction using similarities such as inter-block values within a frame and encodes differences from the original video, and inter prediction performs prediction using similarities such as inter-frame values between blocks And encode the difference from the original video. At the time of intra prediction or inter prediction, P, which is a prediction macro block, is formed based on a reconstructed frame. During intra prediction, P can be formed from samples in a previously encoded current frame, a decoded current frame, or a reconstructed current frame uF _n ′. When P is formed from the reconstructed current frame, unfiltered samples can be used. During inter prediction, P may be formed by performing motion compensation or motion prediction from one or more reference frames. In FIG. 1, it is assumed that the reference frame is F _n−1 ′, which is a previously encoded frame. However, the present invention is not limited to this, and each prediction macroblock may be formed from a previous one frame or two frames that have already been encoded or reconstructed, or a subsequent one frame or two frames.

P, in order to generate a residual (residual_prediction_flag) or differences (difference) macroblock _{D n,} are now subtracted from the macroblock. D _n is transformed (T, transformed) using block transformation, and is quantized (Q, quantized) to generate X. X is a set of encoded coefficients or the like. Encoded coefficients and the like are rearranged and entropy coded, and the entropy coded coefficients and the like form a compressed bitstream together with information necessary for decoding a macroblock. To do. The compressed bit string is sent to a NAL (Network Abstraction Layer) for transmission or storage.

From now on, the reconstruction path will be described. The quantized macroblock coefficient X is decoded to generate a reconstructed frame used for encoding other macroblocks and the like. X is inversely quantized (Q ⁻¹ ) and inversely transformed (T ⁻¹ , Inverse)
Transformed) to generate a difference macroblock D _n ′. The difference macroblock D _n ′ generated in the reconstruction path is not the same as D _n that is the difference macroblock generated in the forward path. Quantization causes losses, so that D _n ′ can be a distorted form of D _n . The predicted macroblock P is added to D _n ′ to generate a reconstructed macroblock uF _n ′. The reconstructed macroblock uF _n ′ may also be a distorted form of the macroblock Fn. A filter can be applied to reduce blocking distortion for uF _n ′, and a reconstructed frame can be formed from a plurality of reconstructed macroblocks to which the filter has been applied.

  FIG. 2 is a block diagram of a H.264 / AVC standard decoder. FIG.

As shown in FIG. 2, the decoder receives a compressed bit string from the NAL. The received data is entropy decoded and rearranged to generate X, which is a set of quantized coefficients and the like. D _n ′ is generated by performing inverse quantization and inverse transform on X, and the decoder uses the same header information (header information) decoded with a bit string as the prediction macroblock generated by the encoder. P is generated. P D _{n 'is} added uF _n' is generated, uF n _'macroblock F _n decoded through the _filter' can be generated.

  Hereinafter, intra prediction will be described.

  When intra prediction is performed on a block (or macroblock), a block (or macroblock) in which a prediction block (or macroblock) P is encoded, a reconstructed block (or macroblock), etc. Can be formed as a base. P is subtracted from the original image, and the difference value obtained by subtracting P is encoded and transmitted. Intra prediction may be performed according to luma prediction mode or chroma prediction mode. Intro prediction may be performed in units of 4 × 4 size sub-blocks or 16 × 16 size macroblocks in the luma prediction mode, and a total of nine additional prediction modes in the 4 × 4 luma prediction mode. However, in the 16 × 16 luma prediction mode, there are a total of four additional prediction modes. The unit in which intra prediction is performed is not limited to sub-blocks or macroblocks, and may be performed in units of various sizes. A unit of pixels on which intra prediction is performed can be referred to as a coding unit (CU) or a prediction unit (PU), and the size of the CU or PU is a sub-block or macroblock as described above. Can be the same size.

  FIG. 3 illustrates an example in which prediction samples are labeled in the 4 × 4 luma prediction mode. As shown in FIG. 3, the prediction block P is calculated based on the samples labeled A to H or I to L.

  FIG. 4 shows nine prediction modes within the 4 × 4 luma prediction mode.

  The encoder can select any one of the nine prediction modes for each block to minimize the difference between the prediction block P and the block to be encoded. The nine prediction modes are as follows.

  1) Mode 0 (vertical): A to D, which are upper samples of the prediction block, are extrapolated vertically.

  2) Mode 1 (horizontal): I to L such as left samples of the prediction block are horizontally extrapolated.

  3) Mode 2 (DC): All samples in P are predicted by the average of A to D and I to L.

  4) Mode 3 (Diagonal Down-Left): Samples in P are interpolated at a 45 ° angle between the lower-left side and the upper-right side.

  5) Mode 4 (Diagonal Down-Right): Samples and the like in P are extrapolated at a 45 ° angle downward to the right.

  6) Mode 5 (Vertical-Right): Samples and the like in P are extrapolated or interpolated at an angle of about 26.6 ° C. from the vertical axis to the right.

  7) Mode 6 (Horizontal-Down): The sample in P is extrapolated at an angle of about 26.6 ° C. downward from the horizontal axis.

  8) Mode 7 (Vertical-Left): Samples and the like in P are extrapolated at an angle of about 26.6 ° C. to the left from the vertical axis.

  9) Mode 8 (Horizontal-Up): Samples in P and the like are interpolated upward at an angle of about 26.6 ° C. from the horizontal axis.

  In FIG. 4, the arrows indicate the direction in which prediction is performed within each mode. On the other hand, for mode 3 to mode 8, the samples in the prediction block are formed from the weighted average of the prediction samples A to H or I to L. For example, in mode 4, it can be predicted with a sample (d = round (B / 4 + C / 2 + D / 4)) located on the upper right side of the prediction block. The encoder calculates sum of absolute errors (SAE) for the prediction blocks generated in each prediction mode, and performs intra prediction in the prediction mode with the smallest SAE.

  FIG. 5 shows an example of a method for applying a filter before intra prediction execution.

  In general, H.W. After applying a filter to samples used in the H.264 / AVC standard, intra prediction is performed. A sample can also be referred to as a reference pixel value. In the example of FIG. 5, it is assumed that the filter is a low-pass filter and the filter is applied only to 8 × 8 blocks.

  Formula 1 is an example of a formula representing a 3-tap filter applied to the reference pixel value.

<Equation 1>
h [Z] = (A + 2 × Z + Q) / 4
h [A] = (Z + 2 × A + B) / 4
. . .
h [P] = (O + 3 × P) / 4
h [Q] = (Z + 2 × Q + R) / 4
. . .
h [X] = (W + 3 × X) / 4

  h [Z] represents a value calculated by applying a filter to Z. As shown in Expression 1, filtering is performed on the reference pixel value by applying the filter coefficient (1, 2, 1), and 9 based on the reference pixel values (h [A] to h [Z]) subjected to the filtering. Intra-prediction is performed using the individual prediction modes. A filter can be applied in the decoding process as in the encoding process.

  When performing the filtering before the intra prediction is performed, the encoding performance may be improved when the filtering is not performed. Accordingly, a method for adaptively applying a filter to perform intra prediction can be proposed.

  FIG. 6 shows an embodiment of a method for performing intra prediction using the proposed adaptive filter.

  As shown in FIG. 6, in step S201, the encoder determines whether or not a filter can be applied to the reference pixel value. When determining whether or not to apply a filter, the encoder can use information of neighboring blocks or determine whether or not to apply a filter by a rate-distortion optimization (RDO) method.

  When determining whether to apply a filter to a reference pixel value using peripheral block information, the prediction mode of the current block is determined based on prediction mode information (MPM: Most Probable Mode) of the peripheral block, and the determined current block Whether to apply a filter to the reference pixel value can be determined according to the prediction mode. For example, assuming that the current block is “C”, the upper block is “A”, and the left block is “B”, the prediction mode of the current block is the same as the prediction mode of “A”. Is determined as the prediction mode of the current block, and when the prediction mode of the current block is the same as the prediction mode of “B”, the prediction mode of “B” can be determined as the prediction mode of the current block. Alternatively, when the prediction mode of the current block is a prediction mode of “A” or another prediction mode excluding the prediction mode of “B”, the prediction mode is encoded and transmitted. Whether to apply the filter to the reference pixel value can be determined according to the prediction mode of the current block determined in this way. Even when the sizes of the current block and the upper block or the left block are different, the prediction mode of the current block can be determined by the prediction mode of the neighboring blocks.

  Alternatively, when determining whether or not a filter can be applied to a reference pixel value using peripheral block information, it is possible to determine whether or not the filter can be applied based on the amount of change in the peripheral reference pixel value. For example, if the reference pixel value to which the filter is to be applied is p [n], the difference value between p [n−1] and p [n + 1], which are neighboring reference pixel values, is calculated, and this is used as a specific threshold value. (Threshold) can be used to determine whether the filter is applicable.

  Alternatively, in addition to the prediction mode of the current block, whether to apply a filter to the reference pixel value can be determined according to the size of the current block. At this time, whether to apply a filter according to the prediction mode of the current block and the size of the current block is designated in advance, and whether to apply the filter adaptively is determined according to the prediction mode or the size.

Table 1 is a table that specifies whether a filter can be applied according to the prediction mode of the current block and the size of the current block.

  Referring to Table 1, “0” means that no filter is applied, and “1” means that a filter is applied. For example, when the current block size is 4 × 4, if the current block prediction mode is 1, the filter is not applied, and if the current block prediction mode is 3, the filter may be applied.

  Also, whether to apply a filter to a reference pixel value can be determined based on whether intra-frame encoding or inter-frame encoding of peripheral blocks is possible. For example, in the case of execution of limited intra-prediction, when a peripheral block is inter-screen encoded, it is encoded between the screens using a value of the peripheral block encoded in the screen. In this case, filtering can not be applied.

  If it is determined to apply the filter to the reference pixel value, the encoder applies the filter to the reference pixel value in step S202. The applied filter may be a commonly used filter. For example, the 3-tap of Equation 1 can be used, or a 2-tap filter can be used. When a 2-tap filter is used, various filter coefficients such as (1/8, 7/8), (2/8, 6/8), (3/8, 5/8) may be used. The reference pixel value to which the filter is applied can be used when applying the filter to other reference pixel values. In addition, when a filter is applied to the reference pixel values, the filter can be applied to all reference pixel values, and the filter can be applied only to some reference pixel values.

  In step S203, the encoder performs intra prediction based on the reference pixel value to which the filter is applied or to which the filter is not applied.

In step S204, the encoder determines whether or not a filter can be applied to the prediction value for each prediction mode predicted by the intra prediction execution for encoding the current block. Here, each prediction mode may be nine prediction modes of 4 × 4 luma prediction mode.
When determining whether or not to apply a filter to a prediction value for each prediction mode, it is possible to determine whether or not to apply a filter by using information on neighboring blocks or by an RDO method.

When determining whether or not to apply a filter to a prediction value using information about a neighboring block, the prediction mode of the current block is determined based on the prediction mode information (MPM) of the neighboring block, and prediction is performed according to the determined prediction mode of the current block. Whether to apply a filter to the value can be determined. For example, assuming that the current block is “C”, the upper block is “A”, and the left block is “B”, the prediction mode of the current block is the same as the prediction mode of “A”. Is determined as the prediction mode of the current block, and when the prediction mode of the current block is the same as the prediction mode of “B”, the prediction mode of “B” can be determined as the prediction mode of the current block. Alternatively, when the prediction mode of the current block is a prediction mode of “A” or another prediction mode excluding the prediction mode of “B”, the prediction mode is encoded and transmitted. At this time, when the prediction mode of the current block is the specific prediction mode (DC or planar), the difference between the reference pixel value and the prediction value may be relatively larger than in other prediction modes. For example, the difference between the reference pixel value and the predicted value in the planar prediction mode may be relatively large compared to other prediction modes. The prediction value in the planar prediction mode is obtained by linearly interpolating each line horizontally.
The average of the first prediction value obtained by interpolation) and the second prediction value obtained by linearly interpolating each column vertically can be calculated. When linearly interpolating horizontally, the value on the right side is the same as the value of the reference pixel value located in the upper right direction (that is, H in FIG. 3). , The reference pixel value is the same as the value located in the lower left direction (that is, I in FIG. 3). Since the predicted value is not immediately obtained from the reference pixel value, the difference between the reference pixel value and the predicted value may be relatively large.
In such a case, a filter can be applied to the predicted value to increase the efficiency of intra prediction. Whether to apply the filter to the prediction value can be determined according to the prediction mode of the current block determined in this way. Even when the sizes of the current block and the upper block or the left block are different, the prediction mode of the current block can be determined by the prediction mode of the neighboring blocks.

  Alternatively, in addition to the prediction mode of the current block, whether or not to apply a filter to the prediction value can be determined according to the size of the current block. At this time, whether to apply a filter according to the prediction mode of the current block and the size of the current block is designated in advance, and whether to apply the filter adaptively is determined according to the prediction mode or the size. Alternatively, whether to apply a filter to a predicted value can be determined depending on whether intra-frame coding or inter-frame coding is performed on neighboring blocks.

  If it is determined to apply a filter to the predicted value, the encoder applies a filter to the predicted value in step S205. Thereby, the prediction of the current block is completed, and the encoder calculates a residual signal and performs entropy coding.

  FIG. 7 shows a case where a filter is applied to a prediction value by the proposed intra prediction execution method using an adaptive filter.

  As shown in FIG. 7, when the prediction mode of the current block is a non-directional mode, the difference between the reference pixel value and the prediction value may be relatively large compared to other prediction modes. The filter can be applied only to the predicted value of the pixel adjacent to the boundary with the peripheral restored reference pixel value. For example, in FIG. 3, it is possible to perform filtering on predicted values corresponding to a1 to a8 and b1, c1, d1, e1, f1, g1, and h1 that are pixels in one row located at the boundary. Or for predicted values corresponding to a1 to a8, b1 to b8 and c1 to c2, d1 to d2, e1 to e2, f1 to f2, g1 to g2, and h1 to h2, which are pixels in two rows located at the boundary Filter can be applied. The filter applied at this time may be a commonly used filter. For example, the 3-tap of Equation 1 can be used, or a 2-tap filter can be used. When a 2-tap filter is used, various filter coefficients such as (1/8, 7/8), (2/8, 6/8), (3/8, 5/8) may be used. Alternatively, any one of a 2-tap filter and a 3-tap filter can be selected and used depending on the position of the pixel.

  In the same way, when the prediction mode of the current block is a prediction mode using reference pixel values corresponding to A to P, such as mode 0, mode 3, or mode 7, between the reference pixel value and the prediction value. A filter can be applied to predicted values corresponding to a1 to a8 having a relatively large difference. Further, when the prediction mode of the current block is a prediction mode using reference pixel values corresponding to Q to X as in mode 1 or mode 8, the difference between the reference pixel value and the prediction value is relatively large a1. , B1, c1, d1, e1, f1, g1, and h1 can be applied to the predicted values.

  Formula 2 is an example of a formula that represents a filter that is applied to the predicted value when a 2-tap filter or a 3-tap filter is selected and used depending on the pixel position.

<Equation 2>
f [a1] = (2 × A + 4 × a1 + 2 × Q) / 8
f [b1] = (2 × B + 6 × b1) / 8
f [c1] = (2 × C + 6 × c1) / 8
. . .
f [a2] = (2 × R + 6 × a2) / 8
f [a3] = (2 × S + 6 × a3) / 8
. . .

  In Formula 2, f [a1] is a value obtained by applying a filter to the predicted value a1, and A and Q represent reference pixel values. Referring to Equation 2, it can be seen that the 3-tap filter is applied to the prediction value of the pixel located at a1, and the 2-tap filter is applied to the prediction values of the remaining pixels.

  Formula 3 is still another example of a formula representing a filter applied to a predicted value when the filter is applied to the predicted value by the proposed intra prediction execution method using an adaptive filter.

<Equation 3>
1. Vertical low-pass filter
v [a1] = (A + 2 × a1 + a2) / 4
v [a2] = (v [a1] + 2 × a2 + a3) / 4
. . .
v [a8] = (v [a7] + 3 × a8) / 4
v [b1] = (B + 2 × b1 + b2) / 4
. . .
2. Horizontal low-pass filter
h [a1] = (Q + 2 × v [a1] + v [b1]) / 4
h [b1] = (h [a1] + 2 × v [b1] + v [c1]) / 4
. . .
h [h1] = (h [g1] + 3 × v [h1]) / 4
h [a2] = (R + 2 × v [a2] + v [b2]) / 4
. . .

  The filter of Equation 3 can be used when applying an adaptive filter based on a method using information of neighboring blocks or an RDO method for each prediction mode. Referring to Equation 3, a low-band filter having a filter coefficient (1, 2, 1) is sequentially executed in two directions, vertical and horizontal. First, a filter is applied in the vertical direction, and a filter is further applied in the horizontal direction based on the value to which the filter is applied. The predicted value to which the filter is applied can be used when applying the filter to other predicted values.

  On the other hand, when the RDO method is applied, the intra prediction execution method described in FIG. 6 can be repeatedly executed.

  FIG. 8 is a block diagram of an encoder and decoder in which an embodiment of the present invention is implemented.

  The encoder 800 includes a processor (810; processor) and a memory (820; memory). The processor 810 implements the proposed functions, processes, and / or methods. The processor 810 determines whether or not to apply the first filter to the reference pixel value based on the information on the neighboring blocks of the current block, and when it is determined to apply the first filter, Applying the first filter, performing intra prediction on the current block based on the reference pixel value, and predicting each prediction mode of the current block predicted by the intra prediction execution based on information on the neighboring blocks When it is determined whether to apply the second filter to another prediction value and it is determined to apply the second filter, the second filter is applied to the prediction value for each prediction mode of the current block. Configured to apply. The memory 829 is connected to the processor 810 and stores various information for driving the processor 810.

  The decoder 900 includes a processor 910 and a memory 920. The processor 910 implements the proposed functions, processes, and / or methods. The processor 910 determines whether or not to apply the first filter to the reference pixel value based on the information of the neighboring blocks of the current block, and when it is determined to apply the first filter, Applying the first filter, performing intra prediction on the current block based on the reference pixel value, and predicting each prediction mode of the current block predicted by the intra prediction execution based on information on the neighboring blocks When it is determined whether to apply the second filter to another prediction value and it is determined to apply the second filter, the second filter is applied to the prediction value for each prediction mode of the current block. Configured to apply. The memory 920 is connected to the processor 910 and stores various information for driving the processor 910.

The processors 810 and 910 are ASIC (application-specific)
integrated circuit), other chipsets, logic circuits, and / or data processing devices. The memories 820 and 920 may include a read-only memory (ROM), a random access memory (RAM), a flash memory, a memory card, a storage medium, and / or other storage medium. When embodiments are implemented in software, the techniques described above can be implemented with modules (processes, functions, etc.) that perform the functions described above. Modules may be stored in memory 820, 920 and executed by processors 810, 910. The memories 820, 920 can be internal or external to the processors 810, 910 and can be coupled to the processors 810, 910 by various well-known means.

  In the exemplary system described above, the method and the like are described as a series of steps or blocks based on a sequence diagram, but the present invention is not limited to the order of steps or the like, and certain steps are described above. Can occur in different steps and in a different order, or simultaneously. Moreover, those skilled in the art will not be aware of the steps shown in the sequence diagrams being exclusive and may include other steps, or one or more steps of the sequence diagrams may not affect the scope of the invention. It will be understood that it can be deleted.

  The above-described embodiments include illustrations of various aspects. Although not all possible combinations for describing various aspects, etc. can be described, those with ordinary knowledge in the art will recognize that other combinations are possible. Accordingly, the present invention includes all other alterations, modifications, and variations that fall within the scope of the following claims.

Claims (3)

Determining whether to apply a first filter to a reference pixel value of a current block;
Performing intra prediction using the reference pixel value induced by applicability of the first filter, and deriving a prediction value for the current block,
In the step of deriving a predicted value for the current block,
Based on whether or not the application condition of the second filter is satisfied, the prediction value for the current block is derived by applying the second filter;
When the second filter is applied, a prediction value for the current block is derived by applying the second filter to a prediction pixel of the current block adjacent to a boundary with the reference pixel value. Either one of the 3-tap filter and the 2-tap filter is selectively applied according to the position of the pixel to which the two filters are applied,
The prediction pixel of the current block adjacent to the boundary with the reference pixel value is the leftmost pixel at the topmost end of the current block, the rightmost pixel at the topmost end of the current block, and the leftmost at the bottommost end of the current block Including pixels,
The 3-tap filter is applied to the leftmost pixel at the uppermost end of the current block, and the 2-tap filter is applied to the rightmost pixel at the uppermost end of the current block;
The first coefficient of the 2-tap filter corresponding to the uppermost rightmost pixel of the current block is the second coefficient of the 2-tap filter corresponding to the pixel adjacent to the uppermost rightmost pixel of the current block. A video decoding method characterized by being three times larger. Determining whether to apply a first filter to a reference pixel value of a current block;
Deriving a predicted value for the current block using the reference pixel value derived by applicability of the first filter;
In the step of deriving a predicted value for the current block,
Based on whether or not the application condition of the second filter is satisfied, the prediction value for the current block is derived by applying the second filter;
When the second filter is applied, a prediction value for the current block is derived by applying the second filter to a prediction pixel of the current block adjacent to a boundary with the reference pixel value. Either one of the 3-tap filter and the 2-tap filter is selectively applied according to the position of the pixel to which the two filters are applied,
The prediction pixel of the current block adjacent to the boundary with the reference pixel value is the leftmost pixel at the topmost end of the current block, the rightmost pixel at the topmost end of the current block, and the leftmost at the bottommost end of the current block Including pixels,
The 3-tap filter is applied to the leftmost pixel at the uppermost end of the current block, and the 2-tap filter is applied to the rightmost pixel at the uppermost end of the current block;
The first coefficient of the 2-tap filter corresponding to the uppermost rightmost pixel of the current block is the second coefficient of the 2-tap filter corresponding to the pixel adjacent to the uppermost rightmost pixel of the current block. A video encoding method characterized in that the video encoding method is three times larger. A computer-readable recording medium storing a computer program for executing the video decoding method according to claim 1 .

Images